The purpose of this project is to develop radial k-space Magnetic Resonance (MR) techniques to improve imaging of cancerous tumors in the abdomen. While MR has been widely embraced as the method of choice for neurological diagnostic imaging applications, it has not received universal acceptance in abdominal applications because the images are frequently contaminated by severe respiratory motion artifacts. In many cases these artifacts render the diagnosis equivocal inasmuch as contrast differences in liver, kidney and adrenal masses can be subtle. This proposal describes a unique approach to enhanced tumor visualization which diminishes the deleterious effects of physiological movement of abdominal structures. Unlike other MRI approaches that rely on corrections to conventional acquisition methods, or on rapid data acquisition for suppression of motion blurring and ghosting, the proposed effort allows conventional TR/TE and SNR choices, and thus provides proven diagnostic imaging contrast characteristics. In the new approach proposed here, radial projection reconstruction (PR) k-space trajectories are shown to have fundamentally superior characteristics relative to those of spinwarp methods with regard to motion artifacts. The plan will carry out a thorough investigation of the tradeoffs of various radial trajectory alternatives, develop a phantom system and quantitative metric for comparison of these techniques, and investigate additional correction schemes which may be employed during reconstruction. The clinical efficacy of these methods will be evaluated in a study of 60 patients with abdominal cancer. The techniques will address the major present limitations of PR imaging: residual streaks from view inconsistencies, off-resonance blurring, motion blurring, and capability for non-axial scan planes. This research will result in the development of superior techniques for MR imaging of abdominal tumors.